timeAverageQueueLength_tendsto_of_scaled_lower_and_upper
ProvedSupporting subproblem for the deterministic continuous-time Little's Law decomposition graph: timeAverageQueueLength_tendsto_of_scaled_lower_and_upper.
Formal statement
import Definitions.Def_queueing_continuous_time
open Filter
open scoped BigOperators Interval Topology
open QueueingLib.LittlesLaw.ContinuousTime
/--
The final squeeze step, stated in the form produced by the proof above. The
upper side comes from arrived sojourns by `t`; the lower side comes from arrived
sojourns by `t / (1 + eps)` after discarding finitely many jobs.
-/
theorem timeAverageQueueLength_tendsto_of_scaled_lower_and_upper
(q : ContinuousSamplePath) (x : ℝ)
(hx : 0 ≤ x)
(hLower : ∀ eps : ℝ, 0 < eps → ∀ delta : ℝ, 0 < delta →
∀ᶠ t in atTop, x / (1 + eps) - delta ≤ timeAverageQueueLength q t)
(hUpper : ∀ delta : ℝ, 0 < delta →
∀ᶠ t in atTop, timeAverageQueueLength q t ≤ x + delta) :
Tendsto (timeAverageQueueLength q) atTop (𝓝 x) := by
-- Depends on a real-analysis squeeze argument with the parameter `eps ↓ 0`.
sorrySource
QueueingLib deterministic continuous-time Little's Law project. Root theorem: general_littles_law, theorem_id 81236938-cf7e-49ad-a078-be9ba5432532.